cognitive approach - methodology Flashcards
laboratory experiment
Experiments carried out in an artificial, controlled setting. The setting is unnatural for PP. The IV is manipulated in some way, usually two conditions+
field experiment
experiment is carried out in a ‘field’ i.e. in the natural setting of the PP e.g. a school for teachers or pupils participating or hospital for medical staff. They are still in control of laboratory experiment included, but these might be difficult to put into place.
experimental hypothesis
this is the alternative to the null hypothesis; that there is a difference as predicted. i.e words from a themed list are better recalled than random words; therefore null hypothesis can be rejected.
directional hypothesis (one-tailed)
the direction of the difference is predicted i.e. if hypothesis is that ‘words from a list with a theme are better recalled than random words on a list’, there is a direction stated - that themed words will be better recalled.
non directional hypothesis (two tailed)
Where the direction of the hypothesis is not predicted ie. there is a difference on the recall of words from a list depending on whether the words are themed or random. This does not five us a direction as to which will be recalled better, just that there will be a difference.
null hypothesis
there is no significant different
operationalisation
making something measurable and clear. The test of recalling themed or random words on list have to be measured; they r recorded
participant design examples
independent measures
repeated measures
matched pairs
independant measures
each group is independent of the other with regard to the people doing the study. One group recalls themed words, another recalls random words
repeated measures
the same pp’s are involved in all the conditions.
matched pairs
different Pp’s in different condition as in independent groups but pp’s are matched in terms of age, gender, experience etc.
order effects
Problems that might occur when the same people are doing all conditions. PP’s responses in the various conditions are affected by the order of conditions to which they were exposed.
practice effects
having done one of the tasks the pp does the next one better because of having had practice.
fatigue effects
having done one task the pp does the next one less well because of being tired/bored
Independent variable
the variable that the researcher changes to see the effect on the dependent variable. e.g. test in recall (DV) the researcher changes the conditions into themed words or random words to see if they have an effect on recall.
dependent variable
the variable that is measured to see the effect of the IV. i.e recall (DV) was found better with themed words than random words (IV’s)
counterbalancing
a way to avoid order effects: alternate the order in which the participant does the conditions
randomising
a way to avoid order effects: chose at random which task the participant will do first. Toss a coin.
Experimental control
an experiment should have control over the many variables that may affect the results of the study. Some variables are easier to control than others.
Extraneous variables
these are any variables that can influence the findings in a study except the independent variable. Sometimes the variable can influence the dependent variable and confound the result.
confounding variables
this is a factor that has not been controlled and has an impact on the findings of the study.
situational variables
variables in the situation such as noise, light, temperature might affect results if they are in different conditions
participant variable
variables in the participant such as mood, hunger, age, gender might affect the results if they are in different conditions
participant variable
variables in the participant such as mood, hunger, age, gender might affect the results if they are in different conditions
ecological validity
how well the study reflects a natural situation. Results can be ‘abnormal’ if experiment carried out in a highly artificial environment. Field and natural experiments have ecological validity as they are conducted in natural surroundings.
construct validity
the extent to which the test measures what it claims to measure e.g. recalling a list of words might be useful measure of episodic memory because a pp may draw on semantic memory and make a good guess
predictive validity
the extent to which results from test can predict future behaviour e.g. test of intelligence can predict future academic success.
internal validity
extent to which outcome of study is a direct result of the independent variable. Internal validity can be ensured by using standardised procedures, controlling for order effects and individuals differences avoiding demand characteristics
external validity
the extent to which the findings apply to other people and situations
experimenter effects
these are things such as tone of researcher’s voice, clothes worn, gender, age - they must be controlled if the experiment is to be truly objective and reliable.
demand characteristics
when characteristics of a questions or features of a study gives clues to the respondent about the answers expected. Questions need to be subtle and hide true intentions.
objectivity
the study is not biased because of the researcher giving their own opinions or influencing the results
reliability
getting the same results again if the study is repeated. To test for reliability, there needs to be replicability.
validity
measuring what you claim to measure, meaning that what you measure is ‘real life’ and not forced because of the study’s methodology.
strength of lab experiment
good control = valid results
cause and effect can be established
strength of field experiment
high ecological validity - take place in natural setting
type I error
null hypothesis wrongly rejected
type II error
null hypothesis wrongly accepted
inter-rater reliability
when more than one pp rates the behaviour and their ratings are compared - if there is an agreement between ratings, its reliable
strength of repeated measures
- pp do all conditions so pp’s variabes (features of pp that might affect results) are controlled. Each pp does all the conditions so any features will affect all conditions and will therefore cancel out
weakness of repeated measures x2
- order effects which may cause bias and drawing wrong conclusions e.g. practice and fatigue effects
- demand characteristics when pp’s responses are affected by guessing what study is about
strength of independent groups
no order effects and less demand characteristics
strength of matched pairs
- pp do all same conditions
- more data gathered as each person does all conditions
weakness of independent groups
more pp are needed because different people do different conditions which means samples take longer to obtain and may be more costly
weakness of matched pairs
- order effects
- demand characteristics
criteria for spearmans rho
test of relationship between two co-variables
at least ordinal data
criteria for man whitney u
test of difference between 2 conditions
ordinal data
independent groups
criteria for wilcoxon
test of difference
ordinal data
repeated measures/matched pairs
criteria for chi sqaured
test of difference
nominal data
independent measures
